Massively Parallel In-Network Compute

2. The method of claim 1, wherein the plurality of switch ports represents a plurality of ingress switch ports; wherein the single result chunk is sent to said each compute node of the plurality of compute nodes via a respective egress switch port in a plurality of egress switch ports connected with the plurality compute nodes.

3. The method of claim 2, wherein the plurality of ingress switch ports is interconnected at least in part through one or more switch fabrics with the plurality of egress switch ports.

4. The method of claim 1, wherein reducing the vector chunks includes performing one or more of: summation, averaging, multiplying, selecting a minimum value, or selecting a maximum value.

5. The method of claim 1, wherein the network switch and the plurality of compute nodes collectively implement a common distributed application.

6. The method of claim 5, wherein the common distributed application includes one or more artificial neural networks.

7. The method of claim 5, wherein the common distributed application represents one or more of: deep learning applications, machine learning applications or artificial intelligence applications.

8. The method of claim 1, wherein an error occurs in processing the vector chunks received from the plurality of compute nodes; wherein a message is sent to each compute nodes in the plurality of compute nodes to inform the error.

9. The method of claim 1, wherein each compute node in the plurality of compute nodes is implemented at least in part with one or more of: central processing units, graphics processing units, tensor processing units, floating point units, hardware accelerators, or other computing processor.

11. The apparatus of claim 10, wherein the plurality of switch ports represents a plurality of ingress switch ports; wherein the single result chunk is sent to said each compute node of the plurality of compute nodes via a respective egress switch port in a plurality of egress switch ports connected with the plurality compute nodes.

12. The apparatus of claim 11, wherein the plurality of ingress switch ports is interconnected at least in part through one or more switch fabrics with the plurality of egress switch ports.

13. The apparatus of claim 10, wherein reducing the vector chunks includes performing one or more of: summation, averaging, multiplying, selecting a minimum value, or selecting a maximum value.

14. The apparatus of claim 10, wherein the network switch and the plurality of compute nodes collectively implement a common distributed application.

15. The apparatus of claim 14, wherein the common distributed application includes one or more artificial neural networks.

16. The apparatus of claim 14, wherein the common distributed application represents one or more of: deep learning applications, machine learning applications or artificial intelligence applications.

17. The apparatus of claim 10, wherein an error occurs in processing the vector chunks received from the plurality of compute nodes; wherein a message is sent to each compute nodes in the plurality of compute nodes to inform the error.

18. The apparatus of claim 10, wherein each compute node in the plurality of compute nodes is implemented at least in part with one or more of: central processing units, graphics processing units, tensor processing units, floating point units, hardware accelerators, or other computing processor.